[SPARK-12298][SQL] Fix infinite loop in DataFrame.sortWithinPartitions

Modifies the String overload to call the Column overload and ensures this is called in a test.

Author: Ankur Dave <ankurdave@gmail.com>

Closes #10271 from ankurdave/SPARK-12298.

sql/core/src/main/scala/org/apache/spark/sql/DataFrame.scala

@@ -609,7 +609,7 @@ class DataFrame private[sql](
    */
   @scala.annotation.varargs
   def sortWithinPartitions(sortCol: String, sortCols: String*): DataFrame = {
-    sortWithinPartitions(sortCol, sortCols : _*)
+    sortWithinPartitions((sortCol +: sortCols).map(Column(_)) : _*)
   }
 
   /**

sql/core/src/test/scala/org/apache/spark/sql/DataFrameSuite.scala

@@ -1090,8 +1090,8 @@ class DataFrameSuite extends QueryTest with SharedSQLContext {
     }
 
     // Distribute into one partition and order by. This partition should contain all the values.
-    val df6 = data.repartition(1, $"a").sortWithinPartitions($"b".asc)
-    // Walk each partition and verify that it is sorted descending and not globally sorted.
+    val df6 = data.repartition(1, $"a").sortWithinPartitions("b")
+    // Walk each partition and verify that it is sorted ascending and not globally sorted.
     df6.rdd.foreachPartition { p =>
       var previousValue: Int = -1
       var allSequential: Boolean = true